March 2010 - Core-other - mailman3.cels.anl.gov

Fwd: Longest Password
by Tina J. Stanton-Piersanti 30 Mar '10

30 Mar '10

ENJOY! > During a recent password audit, > it was found that a blonde was using the following password: > > "MickeyMinniePlutoHueyLouieDeweyDonaldGoofySacramento" > > When asked why she had such a long password, > she said she was told that it had to be at least 8 characters long > and include at least one capital. > >

2 1

Output from "cron" command
by Super-User 29 Mar '10

29 Mar '10

Your "cron" job on mercury /root/bin/daily-snapshot.sh produced the following output: cannot open 'zpool1/avere-ha@Monday': dataset does not exist

1 0

Re: avere testing info
by Max trefonides 29 Mar '10

29 Mar '10

Gents, &cetera, AVERE: This all seems to be working except the mounts from PADS, Ti is working on that, he's put in a request with noc(a)uchicago.edu to put in the exceptions ... As I said in the sync-up the avere is in write Around mode, which means it's essentially another hop to the NFS servers. there are three "vservers" vserver1 : mercury.mcs.anl.gov vserver2 : bbsto05.mcs.anl.gov vserver3 : homenfs.ci.uchicago.edu the avere has a concept of export policies. There is a default policy, which by default is open r/w to the world, (I intend to suggest they make that the inverse). I have edited the default policy on all the vservers to be no access for * I created a policy for the bb nodes that restricts mounting to the 140.221.37.0/24 network I cannot create a policy for pads as you need to be able to moiunt an export before you have anything for which to apply a policy. I have turned on the "High Availability" option, whih essentially means that there is an NFS export that both of the avere nodes write heartbeat data into. The export is on mercury at zpool1/avere-ha and given it a 10GB quota so if something does go wayhire, (it shouldn't) it won't fill up mercury's filesystem. In addition to throughput testing we need to test CIFS exporting. I think that's it, in an emergency or if I've set something up with the crazy and you don't understand my genius-ness I will be watching my personal e-mail: max(a)trefonides.com but I am going to try to ignore mcs mail ... also I am going to try to forward my home phone and my cell phone to a voip number, that may or may not work, but I will get an attachment in e-mail if you leave me v-mail. Later! -- I hope I get to meet Monty Python -- max On Mar 26, 2010, at 3:52 PM, Max trefonides wrote: > I have done some handing off to Ti. > > The tests should be long reads and writes of large files and tiny files, with timing. The avere should be > > Test: > > mounting the backend server directly > through the avere in write around mode > through the avere in write through mode > through the avere in write aback mode > > My iozone test results are in: > > http://www.mcs.anl.gov/~maxadam/avere_testing/ > > Follow the bouncing ball ... > > The naming scheme for the dirs is pretty self explanitory, but here's a key for the bit thats not: > wa == "Write Around" > wt == "Write Through" > wb == "Write Back" > > I set up index.html symlinks in most of the dirs, but I think I missed some. The graphs are in the file with "-report" in it's name, (the columns in a browser tend to cut off the end of the filename, so it's not obvious which one is the html file) > > The avere network is thus: > > NOTE "-sn" delineates server net interface addresses. > > avere-mgt-sn.anl.gov > 140.221.8.192 > avere-cluster-sn.anl.gov = NFS backend server facing cluster net (NFS servers export to this address) > 140.221.8.190 > 140.221.8.191 > > avere-sn.mcs.anl.gov = client facing cluster net (mcs clients mount from this address) > 140.221.8.211 > 140.221.8.212 > > bb-avere-sn.mcs.anl.gov = client facing cluster for breadboard (bb clients mount from this address) > 140.221.8.223 > 140.221.8.224 > > pads-avere-sn.mcs.anl.gov = client facing cluster for breadboard (pads clients mount from this address) > 140.221.8.236 > 140.221.8.237 > > > these were some of the big-ish blocks of IPs that were available, I chose to put each interface in different vlocks in case there was a reason to add more IPs to them. The avere uses contiguous IPs , the interface asks for a beginning address and an ending address. > > 8.190 - 8.203 = 14 > 8.206 - 8.209 = 4 > 8.211 - 8.217 = 7 > 8.223 - 8.231 = 9 > 8.236 - 8.249 = 14 > 8.251 - 8.254 >

1 0

where's Linda 3/29-30
by Linda Winkler 29 Mar '10

29 Mar '10

vacation days lw

1 0

Taking Monday morning off
by Craig Stacey 28 Mar '10

28 Mar '10

We found an apartment, and I need to take tomorrow morning to get us moved out of the hotel and into the apartment. It was supposed to happen today, but the bits and pieces in making that happen went far slower than expected. I may need to take the whole day, we'll see. For now let's just call it a half day. == Craig

1 0

Output from "cron" command
by Super-User 28 Mar '10

28 Mar '10

Your "cron" job on mercury /usr/lib/fs/nfs/nfsfind produced the following output: find: stat() error /exports/proj-fischer00/fischer/pc/00000001.JPG: Value too large for defined data type find: stat() error /exports/proj-fischer00/fischer/pc/00000002.JPG: Value too large for defined data type find: stat() error /exports/proj-fischer00/fischer/pc/My Music/00000002.JPG: Value too large for defined data type find: stat() error /exports/proj-fischer00/fischer/pc/My Music/00000001.JPG: Value too large for defined data type find: stat() error /exports/proj-fischer00/fischer/pc/My Pictures/RR/DMIR_6/DMIR222a.JPG: Value too large for defined data type find: stat() error /exports/proj-fischer00/fischer/pc/My Pictures/RR/DMIR_6/DMIR222b.JPG: Value too large for defined data type find: stat() error /exports/proj-fischer00/fischer/pc2/00000001.JPG: Value too large for defined data type find: stat() error /exports/proj-fischer00/fischer/pc2/nek5n/doc/movies: Value too large for defined data type find: stat() error /exports/proj-fischer00/fischer/pc2/My Music/00000002.JPG: Value too large for defined data type find: stat() error /exports/proj-fischer00/fischer/pc2/My Music/00000001.JPG: Value too large for defined data type find: stat() error /exports/proj-fischer00/fischer/pc2/My Pictures/RR/DMIR_6/DMIR222b.JPG: Value too large for defined data type find: stat() error /exports/proj-fischer00/fischer/pc2/My Pictures/RR/DMIR_6/DMIR222a.JPG: Value too large for defined data type find: stat() error /exports/proj-fischer00/fischer/pc2/00000002.JPG: Value too large for defined data type find: stat() error /exports/shared/nt-homes001/lusk/mpich/examples/contribtest/datatypes/bc6/bcr0c1.err: Value too large for defined data type find: stat() error /exports/shared/nt-homes001/lusk/mpich/examples/contribtest/datatypes/bc1/bcr0c1.err: Value too large for defined data type find: stat() error /exports/shared/nt-homes001/lusk/mpich/examples/contribtest/datatypes/bc5/bcr0c1.err: Value too large for defined data type find: stat() error /exports/shared/nt-homes001/lusk/mpich/examples/contribtest/datatypes/bc10/bcr0c1.err: Value too large for defined data type find: stat() error /exports/shared/nt-homes001/lusk/mpich/examples/contribtest/datatypes/bc2/bcr0c1.err: Value too large for defined data type find: stat() error /exports/shared/nt-homes001/lusk/mpich/examples/contribtest/datatypes/bc8/bcr0c1.err: Value too large for defined data type find: stat() error /exports/shared/nt-homes001/lusk/mpich/examples/contribtest/datatypes/bc0/bcr0c1.err: Value too large for defined data type find: stat() error /exports/shared/nt-homes001/lusk/mpich/examples/contribtest/datatypes/bc7/bcr0c1.err: Value too large for defined data type find: stat() error /exports/shared/nt-homes001/lusk/mpich/examples/contribtest/datatypes/bc9/bcr0c1.err: Value too large for defined data type find: stat() error /exports/shared/nt-homes001/lusk/mpich/examples/contribtest/datatypes/bc3/bcr0c1.err: Value too large for defined data type find: stat() error /exports/shared/nt-homes001/lusk/mpich/examples/contribtest/datatypes/bc11/bcr0c1.err: Value too large for defined data type find: stat() error /exports/shared/nt-homes001/lusk/mpich/examples/contribtest/datatypes/bc4/bcr0c1.err: Value too large for defined data type find: stat() error /exports/shared/nt-homes001/lusk/mpich/examples/perftest/long.err: Value too large for defined data type find: stat() error /exports/shared/nt-homes001/lusk/mpich/examples/perftest/lrtestf.c: Value too large for defined data type find: stat() error /exports/shared/nt-homes001/lusk/mpich/mpid/ch_nexus/nexus-rm/quickthreads/md/null.README: Value too large for defined data type find: stat() error /exports/shared/nt-homes001/lusk/mpich/mpid/ch_nexus/nexus-rm/data_conversion/stamp-h.in: Value too large for defined data type find: stat() error /exports/shared/nt-homes001/lusk/mpich/mpid/ch_nexus/nexus-rm/ports0/stamp-h.in: Value too large for defined data type find: stat() error /exports/shared/nt-homes001/lusk/mpich/mpid/ch_nexus/nexus-rm/ports0/p0_th_null_i.h: Value too large for defined data type find: stat() error /exports/shared/nt-homes001/lusk/mpich/mpid/ch_nexus/nexus-rm/ports0/p0_th_null.h: Value too large for defined data type find: stat() error /exports/shared/nt-homes001/lusk/mpich/mpid/ch_nexus/nexus-4.0b4/config/CONFIG: Value too large for defined data type

1 0

[CELS Sysadmins] Geekbeer
by Craig Stacey 26 Mar '10

26 Mar '10

I have to make an Ikea run at some point this weekend. If we beer at Gordon Biersch or Auld Dubliner, I can do that with ease. == Craig _______________________________________________ This is the CELS Sysadmins mailing list. It is sent to memebers of: core-admins(a)mcs.anl.gov, hpc-admins(a)mcs.anl.gov, systems(a)alcf.anl.gov, aig(a)alcf.anl.gov. Do not subscribe directly to this list. Thanks.

3 4

Re: the pager is getting stuff from 221 still
by Craig Stacey 26 Mar '10

26 Mar '10

We need the pagers, we just don't care about 221 environmentals anymore. Judy Stickels <stickels(a)mcs.anl.gov> wrote: Do you still need the pagers? If not, let's turn them in. --Judy On Mar 26, 2010, at 11:12 AM, Craig Stacey wrote: > Judy, can we get those turned off now? > > On Mar 26, 2010, at 9:45 AM, Rick Bradshaw wrote: > >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> I have been forgetting to send this mail, but the pager is still >> getting alerts from PFS about building 221 stuff. Who would we >> contact >> about getting that changed? >> >> Rick >> -----BEGIN PGP SIGNATURE----- >> Version: GnuPG v1.4.9 (GNU/Linux) >> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org >> >> iEYEARECAAYFAkusyKYACgkQdavihOKkCCFW2QCdGi/1WXyZOvUaYMVKSHCI/qjE >> HhcAoIGLwS4zrbXhr9Qdas7YMuFAuDAU >> =sDM9 >> -----END PGP SIGNATURE----- >> >> >> > > == > Craig > > > Judy Stickels Manager, TCS Building Operations MCS/240, Rm. 3142 Argonne National Laboratory 9700 South Cass Avenue Argonne, IL 60439 630-252-3369 (voice) stickels(a)mcs.anl.gov

2 3

Re: Lol
by Craig Stacey 26 Mar '10

26 Mar '10

Take a picture! bradshaw(a)mcs.anl.gov wrote: This is Jolene they dililated ricks eyes he not coming back to woork can't see

1 0

Lol
by bradshaw＠mcs.anl.gov 26 Mar '10

26 Mar '10

This is Jolene they dililated ricks eyes he not coming back to woork can't see

1 0